ltcai 4.6.0

Lattice AI — local-first Living Brain workspace: conversation, durable memory, hybrid search, real agent/workflow runtimes, advanced graph exploration, and portable encrypted brain archives

Lattice AI

Lattice AI v4.6.0 is a local-first Living Brain desktop workspace. It runs as a Tauri desktop app with a localhost FastAPI sidecar, stores the user's brain locally by default, and makes the Brain plus conversation the primary product experience.

This README describes the v4.6.0 release-preparation tree, which preserves the v4.4.0 physical lattice_brain extraction, v4.5.0 capability recovery, and v4.5.1 product shell while repositioning the graph as a deep exploration layer. First launch now follows a Brain-first product flow: Login, friendly environment analysis, recommended models, guided install/download/validate/load, then the living Brain conversation. Memories, knowledge, relationships, and the advanced graph are progressively disclosed beneath the Brain. External package registries are owner-published; the badges above link to package pages and may show the most recently published owner-controlled registry version, which can lag behind the GitHub Release.

What Was Verified

• Desktop app starts and serves the React/Vite product shell from the FastAPI sidecar.
• First launch opens to Login only, then guides environment analysis, recommendations, install/download/validate/load, and finally the Brain.
• After model loading, the home route opens to a living Brain presence plus durable conversation, with the Brain always visible during the primary chat flow.
• The Brain presence reacts to conversation, recall, model readiness, and agent/workflow activity with neural movement, memory pulses, and status changes.
• Memories, Knowledge, Relationships, and Graph are ordered as progressive disclosure layers; the graph is no longer the product's first impression.
• Capture uploads a document through /upload/document and shows indexed documents from the backend.
• Act exposes workflow create/run surfaces and agent runtime status without presenting simulation as real success. When no LLM-backed model is loaded, deterministic model-free agent simulation is reported honestly and does not call a model.
• Model setup recommends a short ranked list for the current computer instead of exposing a catalog or runtime internals.
• Gemma 4 MLX models are checked against their local config.json before load: Gemma 4 12B gemma4_unified now shows Runtime update needed when the installed MLX-VLM lacks mlx_vlm.models.gemma4_unified, while Gemma 4 26B A4B stays on the working gemma4 MLX-VLM path.
• System exposes storage, backup health, archive, Brain Network, device identity, and admin status through real APIs.
• Backup, restore dry-run, archive verify, archive import dry-run, and .latticebrain portability flows were exercised.

Product Tour

Desktop Startup

The v4.6.0 DMG build launches a visible Tauri app, starts the FastAPI sidecar on localhost, and shuts that sidecar down on normal macOS quit.

Evidence:

• Health log: output/audits/v4.3.2-rc/logs/desktop-sidecar-health-after-shutdown-fix.json
• Shutdown log: output/audits/v4.3.2-rc/logs/desktop-shutdown-after-fix.txt

Living Brain

Brain opens on a living presence and conversation. The visual Brain is not a logo or static decoration: it changes state while the user types, when memories are recalled, while a response streams, and when agent/workflow activity is reported.

The advanced graph still exists, but it is intentionally opened from the Graph layer after the user moves through Brain, Memories, Knowledge, and Relationships.

Conversation

Conversation uses backend conversation, model, and memory APIs. If no model is loaded, the UI shows a ready/unavailable state instead of fabricating an answer while still keeping nearby memory and source signals visible.

Capture

Capture sends files through the backend upload/ingestion path and shows indexed documents returned by the backend. The screenshot below includes a note uploaded during release-prep evidence capture.

Upload evidence: output/audits/v4.3.2-rc/logs/readme-upload-note.json

Act

Act exposes agents, workflows, approvals, hooks, and runtime status through existing backend APIs. The audited workflow path created a real workflow record; agent runtime simulation remains honestly unavailable as product success when no LLM-backed model is loaded.

Library

Library shows models, skills, MCP, and marketplace/configuration state with availability reported from runtime APIs. Optional model runtimes are not treated as loaded unless the runtime is actually available. Model setup now follows the explicit Environment Analysis -> Recommended Models -> Install -> Download Progress -> Validate -> Load -> Ready path, with download/install consent kept visible.

System

System exposes account/workspace status, storage mode, backup health, archive operations, device identity, Brain Network, and admin hardening state. External integrations remain opt-in.

Backup / Restore

Backup and restore flows are backed by the knowledge-graph portability APIs. The retained v4.3.2 product audit exercised backup health, backup creation, restore dry-run, and archive restore surfaces.

.latticebrain Portability

The portable brain format is an encrypted .latticebrain archive. The audited System flow exercised archive export, inspect, verify, import dry-run, and restore dry-run/restore controls through FastAPI.

Archive evidence:

• Create: output/audits/v4.3.2-rc/logs/archive-create.json
• Verify: output/audits/v4.3.2-rc/logs/archive-verify.json
• Import dry-run: output/audits/v4.3.2-rc/logs/archive-import-dry-run.json

Architecture At A Glance

• Desktop shell: Tauri 2 primary shell; Electron remains fallback-only.
• Frontend: React, TypeScript, Vite, TanStack Query, Zustand, Cytoscape.js, React Flow, Tailwind/shadcn-style primitives, generated OpenAPI client.
• Backend: FastAPI on localhost is the source of truth for the UI.
• Brain Core: independent Python package lattice_brain (graph, memory, context, conversations, ingestion, agent/hook runtime, workflow, portability, storage) physically hosted in the package, imported by FastAPI, CLI, tests, and future tools, and guaranteed by tests to never import latticeai.
• Storage: StorageEngine abstraction with SQLite default and optional PostgreSQL/pgvector scale mode.
• Portability: encrypted .latticebrain archive plus backup/restore and migration tooling.
• Privacy: local-first by default; cloud models, Telegram, Brain Network, Docker, model downloads, and update checks require explicit opt-in paths.

See ARCHITECTURE.md for the detailed v4 architecture. v4.6.0 does not redesign the v4.4.0 Brain Core extraction, StorageEngine, FastAPI, Tauri, backup/restore, or portability architecture.

Installation And Release Artifacts

Validated v4.6.0 artifacts are produced from the Living Brain tree:

• dist/ltcai-4.6.0-py3-none-any.whl
• dist/ltcai-4.6.0.tar.gz
• ltcai-4.6.0.tgz
• dist/ltcai-4.6.0.vsix
• src-tauri/target/release/bundle/dmg/Lattice AI_4.6.0_aarch64.dmg

For a public release, attach only those exact artifacts to the GitHub Release. Package-registry publishing is reserved for the owner.

Local Development

npm install
npm run dev

Open the local app:

http://127.0.0.1:4825/app

Build the release artifacts:

npm run release:artifacts
npm run release:validate

Run the main validation set:

npm run check:python
node scripts/run_python.mjs -m ruff check .
npm run lint
npm run typecheck
npm run test:unit
LTCAI_TEST_BASE_URL=http://127.0.0.1:4932 npm run test:integration
npm run test:visual
npm run desktop:tauri:check
node scripts/run_python.mjs scripts/wheel_smoke.py --wheel dist/ltcai-X.Y.Z-py3-none-any.whl
npm pack --dry-run

Known Limitations

• External package registries are owner-published and can lag behind the GitHub Release.
• PostgreSQL/pgvector is optional scale mode and is not required for default local use.
• Docker is never auto-started by default.
• Model downloads and cloud model calls require explicit user action/consent.
• Conversation does not fabricate answers when no model is loaded.
• Historical artifacts may remain in dist/; release uploads must use exact v4.6.0 filenames for this release.

Release History

Version	Theme
4.6.0	Living Brain Experience: made Brain plus conversation the home product, added an animated living Brain presence, and moved graph exploration to the deepest intentional layer
4.5.1	Product Reimagining RC: replaced the desktop shell, navigation model, onboarding journey, first-viewport hierarchy, and visual system while preserving capabilities and local-first architecture
4.5.0	Product Experience Recovery RC: restored first-run setup, workspace/model onboarding, explicit model install/download/validate/load flow, Gemma 4 runtime compatibility gating, Basic-mode polish, and graph discoverability
4.4.0	Brain Engine Extraction: Brain Core physically moved into lattice_brain (graph/memory/context/conversation/ingestion/runtime/workflow/portability), latticeai paths reduced to compatibility shims, isolation tests guaranteeing no latticeai imports
4.3.3	Dead-Code Cleanup Release: post-audit cleanup, architecture documentation correction, Vercel/static-docs readiness, README badge restoration, exact-current artifacts
4.3.2	Product Polish & Graph UX Overhaul RC: evidence-based README, graph UX, structured product state, archive UX, desktop sidecar cleanup, publishing readiness
4.3.1	End-User Audit Repair RC: desktop sidecar startup, npm clean install, default-off downloads, honest agent/workflow states
4.3.0	Portability & Product Hardening RC: encrypted .latticebrain archives, backup/restore hardening, local-only startup guards
4.2.0	Brain Core & Storage Rebuild: independent lattice_brain, StorageEngine abstraction, SQLite default, opt-in Postgres/pgvector
4.1.0	Frontend & Desktop Rebuild: React/Vite SPA, Tauri shell, generated OpenAPI client
4.0.1	Digital Brain maintenance: durable async runs, identity/workspace state, /app parity
4.0.0	Digital Brain Platform foundation
3.0.0	v3 local-first AI workspace platform

Current Documentation

• ARCHITECTURE.md - v4 architecture.
• FEATURE_STATUS.md - current feature status and historical status ledger.
• RELEASE_NOTES.md - current release notes index.
• RELEASE_NOTES_v4.6.0.md - v4.6.0 Living Brain release notes.
• RELEASE_NOTES_v4.5.1.md - v4.5.1 RC release notes.
• RELEASE_NOTES_v4.5.0.md - v4.5.0 RC release notes.
• RELEASE_NOTES_v4.4.0.md - v4.4.0 release notes.
• RELEASE_NOTES_v4.3.3.md - v4.3.3 release notes.
• RELEASE.md - release checklist and exact artifact guidance.
• SECURITY.md - security posture.
• docs/CHANGELOG.md - changelog.
• docs/V4_3_2_DEADCODE_AUDIT_REPORT.md - v4.3.3 cleanup basis.
• docs/V4_3_3_VALIDATION_REPORT.md - v4.3.3 validation report.
• docs/V4_5_0_PRODUCT_EXPERIENCE_RECOVERY_REPORT.md - v4.5.0 product recovery report.
• docs/V4_5_0_VALIDATION_REPORT.md - v4.5.0 validation report.
• docs/V4_5_1_PRODUCT_REIMAGINING_REPORT.md - v4.5.1 product reimagining report.
• docs/V4_5_1_UX_REPORT.md - v4.5.1 UX report.
• docs/V4_5_1_VISUAL_DESIGN_REPORT.md - v4.5.1 visual design report.
• docs/V4_5_1_NAVIGATION_REPORT.md - v4.5.1 navigation report.
• docs/V4_5_1_ONBOARDING_REPORT.md - v4.5.1 onboarding report.
• docs/V4_5_1_MODEL_EXPERIENCE_REPORT.md - v4.5.1 model experience report.
• docs/V4_5_1_GRAPH_EXPERIENCE_REPORT.md - v4.5.1 graph experience report.
• docs/V4_5_1_VALIDATION_REPORT.md - v4.5.1 validation report.
• docs/V4_5_1_RC_ARTIFACTS.md - v4.5.1 screenshot, GIF, and RC artifact manifest.
• docs/V4_6_0_LIVING_BRAIN_EXPERIENCE_REPORT.md - v4.6.0 Living Brain design and validation notes.
• docs/V4_3_2_GRAPH_UX_REPORT.md - graph UX report.
• docs/V4_3_2_PRODUCT_POLISH_REPORT.md - product polish report.
• docs/V4_3_2_SELF_AUDIT_REPORT.md - self-audit evidence.
• docs/V4_3_2_VALIDATION_REPORT.md - RC validation report.
• docs/V4_3_2_DOCUMENTATION_CLEANUP_REPORT.md - release-prep documentation cleanup report.
• docs/V4_3_2_GITHUB_VERCEL_CHECK_REPORT.md - GitHub/Vercel readiness report.
• docs/V4_3_2_INDEPENDENT_AUDIT_PACKAGE.md - package for an independent reviewer.
• docs/V4_3_2_DEADCODE_AUDIT_REPORT.md - independent dead-code, architecture, and runtime audit.
• docs/V4_DIGITAL_BRAIN_RECOVERY.md - transformation recovery file.

License

MIT. See LICENSE.